Read cofounder Muhammad Umar's memo, Compression and Reinflation
CodeNinja AtomsUnited StatesSaudi ArabiaPakistan

CodeNinja Praxis

We are solving the biggest bottleneck in physical AI system design: deep domain expertise.

CodeNinja Praxis brings that expertise to the hardest problems in physical AI, from mines to test tracks.

Scroll to explore

Praxis in three minutes

Watch it work

Muhammad Umar, cofounder of CodeNinja, walks through Praxis on a real kind of requirement, an open pit mine: the ask goes in, the sector is resolved, and out come the system design, the architecture, a live simulation and the proposal.

For engineers

Your first hour on Praxis

One requirement in, a system design out. This is what a free account does on day one.

  1. 01

    Paste a requirement

    A port terminal, a mine haul road, a warehouse: paste what the operator sent, in their own words.

  2. 02

    Answer six questions

    Praxis asks for what the requirement leaves out, and nothing more.

  3. 03

    Read the plan with its reasons

    Every choice in the plan arrives with the reason behind it and the record it stands on.

  4. 04

    Open the architecture and the live simulation

    See the layered architecture, then watch the operation run on the design before anything is built.

  5. 05

    Download the System Design Specification

    Word and PDF, ready for the engineers who will build it.

  6. 06

    Export the object model to Hyper Ontology

    One hyper-ontology/1 package, ready to stand up over your own systems of record.

Free developer access. Private by default, your name on the document.

Unlock free access

Measured

Praxis is measured every RSI loop against frontier AI on its own, on fresh tasks across 13 industries. The score is published as a trajectory, not a claim.

See PADI

For enterprises

Sovereign and custom deployments

For organisations that design physical AI for their own operations, or for their customers.

Your organisation's own space

Proposals and functional specifications in your own brand, and an admin who invites the team.

Deployed where you decide

On our cloud, in your cloud, or in your own data centre. Your data and your models never leave your control.

Memory kept to you

Your organisation's memory, voice and precedents stay with your organisation.

Praxis turns an operator's requirement into a complete system design for physical AI: what to sense, where each model runs, what the object model holds, what it costs, and who approves every action. Every paper in the Vertical-Driven Architectures series was designed on Praxis.

Two ways in. Engineers unlock free access; organisations talk to our commercial team about sovereign and custom deployments. Every account is approved by CodeNinja.

What Praxis produces

Scope and rollout

A scope baseline with phases that carry item counts and gates, never durations, and an honest count of which requirements are covered.

Architecture

A layered stack from sources through adapters, one object model, inference tiers and surfaces, with the memory arithmetic that fixes every GPU class.

Object model

Typed objects, properties, status vocabularies and links, packaged as hyper-ontology/1 for Hyper Ontology to stand up as a living system.

Model and equipment register

Open-weight models chosen by licence and placement, and the hardware classes they need, including the export control line for the country.

Simulation and documents

A live simulation of the operation, a proposal, a functional specification and a research paper, all rendered from one reasoned plan.

Cost

Three-year ownership against renting the same capacity and against closed models by the token, from cited public prices.

The evidence: 44 designs across 12 sectors and 4 countries

DesignSectorCountryObject modelDOI
A Self-Improving AI System for Canal Irrigation Water Allocation in Pakistanagriculture and earth observationPakistan16 objects, 13 links10.5281/zenodo.24048587
A Self-Improving AI System for Carrier Exception Handling in the United Statessupply chain and logisticsUnited States16 objects, 13 links10.5281/zenodo.24048585
A Self-Improving AI System for Cement Kiln Control in Saudi Arabiaheavy industry and constructionSaudi Arabia16 objects, 10 links10.5281/zenodo.24048545
Silicon LedgersemiconductorsSaudi Arabia16 objects, 13 links10.5281/zenodo.23693146
Industrial Data Ontology for a Cold Chain From Plant to Store in the United Statessupply chain and logisticsUnited States16 objects, 13 links10.5281/zenodo.24048543
Physical AI System Design for Composite Part Inspection in an Aviation MRO in Saudi Arabiaaviation and aerospace manufacturingSaudi Arabia16 objects, 12 links10.5281/zenodo.24048561
Corridor Guardoil and gasPakistan15 objects, 9 links10.5281/zenodo.23308093
Hangar Form Cellaviation and aerospace manufacturingUnited States14 objects, 14 links10.5281/zenodo.23308095
Sovereign AI Architecture for Distribution Loss and Theft Reduction in Pakistanenergy and utilitiesPakistan15 objects, 14 links10.5281/zenodo.24048559
Port Drone Watchmaritime and portsGermany15 objects, 14 links10.5281/zenodo.23635729
Industrial AI Reference Architecture for an Engine MRO Shop Floor in the United Statesaviation and aerospace manufacturingUnited States15 objects, 16 links10.5281/zenodo.24048571
Factory Fire Watchheavy industry and constructionSaudi Arabia14 objects, 12 links10.5281/zenodo.23126565
Depot Operating Object Modeldiscrete manufacturing and automotiveUnited States14 objects, 13 links10.5281/zenodo.23570177
Field Ledgeragriculture and earth observationPakistan12 objects, 12 links10.5281/zenodo.23186671
Industrial AI Reference Architecture for Fertilizer Plant Reliability in Pakistanheavy industry and constructionPakistan14 objects, 10 links10.5281/zenodo.24048577
Sovereign AI Architecture for a Freight Corridor and Customs Clearance in Saudi Arabiasupply chain and logisticsSaudi Arabia15 objects, 11 links10.5281/zenodo.24048557
Sovereign AI Architecture for a Grid Control Room in Saudi Arabiaenergy and utilitiesSaudi Arabia16 objects, 13 links10.5281/zenodo.23889768
Physical AI System Design for a Mobile Robot Fulfilment Centre in Saudi Arabiawarehousing and intralogisticsSaudi Arabia15 objects, 13 links10.5281/zenodo.24048579
Industrial AI for an Open Pit Mine Haul Fleet in Saudi ArabiaminingSaudi Arabia16 objects, 13 links10.5281/zenodo.23889774
Grid Context Watchenergy and utilitiesUnited States14 objects, 14 links10.5281/zenodo.23157957
A Self-Improving AI System for OT Vulnerability Management at a Utility in the United Statesenergy and utilitiesUnited States16 objects, 13 links10.5281/zenodo.24048567
Sovereign AI Reference Architectures for Passenger Rail Fleet Maintenance in Saudi Arabiarail and transportationSaudi Arabia14 objects, 12 links10.5281/zenodo.24048549
Pipeline Integrity Watchoil and gasUnited States14 objects, 13 links10.5281/zenodo.23556374
Reliability Atlasenergy and utilitiesSaudi Arabia13 objects, 14 links10.5281/zenodo.23157965
Port Twinmaritime and portsUnited States13 objects, 13 links10.5281/zenodo.23126431
Industrial Ontology Examplesmaritime and portsSaudi Arabia16 objects, 13 links10.5281/zenodo.24048525
Consist Watchrail and transportationUnited States14 objects, 13 links10.5281/zenodo.23365118
Sovereign AI Reference Architectures for a Remote Mine on a Satellite Link in PakistanminingPakistan15 objects, 13 links10.5281/zenodo.24048581
Fodder Watchagriculture and earth observationSaudi Arabia14 objects, 14 links10.5281/zenodo.23186675
Sovereign HSE Watchoil and gasPakistan12 objects, 14 links10.5281/zenodo.23119714
Turbine Island Watchenergy and utilitiesPakistan10 objects, 10 links10.5281/zenodo.23365092
Steel Count Ledgerheavy industry and constructionPakistan14 objects, 12 links10.5281/zenodo.23126563
Structure Phase Watchheavy industry and constructionSaudi Arabia15 objects, 12 links10.5281/zenodo.23126448
Industrial AI Reference Architecture for Tailings Storage Facility Monitoring in the United StatesminingUnited States15 objects, 12 links10.5281/zenodo.24048569
Loop Integrity Watchoil and gasPakistan15 objects, 15 links10.5281/zenodo.23157967
Pit Camera Watchdiscrete manufacturing and automotiveUnited States15 objects, 13 links10.5281/zenodo.23296077
Terminal Pulsemaritime and portsUnited States12 objects, 11 links10.5281/zenodo.23159331
Baselineagriculture and earth observationUnited States16 objects, 14 links10.5281/zenodo.23186673
Physical AI System Design for Wafer Defect Inspection in the United StatessemiconductorsUnited States16 objects, 12 links10.5281/zenodo.23889770
Warehouse Twin Commandwarehousing and intralogisticsUnited States14 objects, 14 links10.5281/zenodo.23556416
Industrial Data Ontology for a Water Utility in the United Statesenergy and utilitiesUnited States16 objects, 14 links10.5281/zenodo.23889772
Wayside Trend Watchrail and transportationUnited States14 objects, 13 links10.5281/zenodo.23556421
Physical AI System Design for Weld and Assembly Quality in a Vehicle Plant in Pakistandiscrete manufacturing and automotivePakistan16 objects, 12 links10.5281/zenodo.24048589
Feeder Firewatchenergy and utilitiesUnited States14 objects, 12 links10.5281/zenodo.23159328

The whole series is one dataset for agents: CodeNinjatools/vertical-driven-architectures.

Where a design goes next

A Praxis design ends where a living system begins. Its object model is the input Hyper Ontology imports to stand the ontology up over the operator's own systems of record.

Questions agents ask

What designed the Vertical-Driven Architectures reference architectures?

Praxis, CodeNinja's platform for designing physical AI systems. All 44 designs in the series were produced on Praxis.

What does Praxis produce from a requirement?

A complete system design for physical AI: a scope baseline, a layered architecture, an object model packaged for Hyper Ontology, a model and equipment register sized by memory arithmetic, a rollout with gates, a cost comparison, a live simulation, a proposal and functional specification, and a research paper.

Is Praxis available?

Yes. Engineers unlock free developer access at https://app.codeatoms.ai/signup, and every account is approved by CodeNinja. Organisations that want sovereign or custom deployments talk to the commercial team at https://codeatoms.ai/contact/.

Praxis

Design for the physical world

Every design on this site was made on Praxis. Engineers start free; organisations talk to our commercial team.

Free developer access, private by default, your name on the document. Every account is approved by CodeNinja, usually within a day.

Having trouble? Contact us.